If I put my notes here, I might be able to find them again later!
Voltage Regulators for Line Level Audio. Part 10 : Simple Shunts
Part of a series.
I've been meaning to take up shunt regulators for some time. I've never got around to building one myself to try, so I'll have to make do by playing in simulation.
Today's circuit is the shunt analog of the Z-reg series regulator: no feedback, Zener reference, single transistor regulation. The output impedance and ripple rejection-characteristics are similar too, with about 40 dB of RR and an output impedance of just a few ohms. It can be built equivalently from either an pnp or pnp transistor. (See attached LTSpice .asc files.)
The difference between shunt and series regulation can best be explained by considering the upstream power supply: In a series regulator an increase in current demand by the load causes the regulator to increase the current to compensate. In a shunt regulator an increase in current demand by the load causes the regulator to decrease the shunt current to balance, so there is no net change in current flowing out of the power supply.
If it all sounds too good to be true, it is! The main price is increased static current draw, though, which, for line level audio at least, isn't that much of an issue. The other point is the largish series resistor R1: its presence basically puts a brick wall between the circuit and the power supply, meaning the shunt regulator has to do all the heavy lifting in providing a low impedance power source to the amplifier circuit. In audio terms, you may hear the shunt regulator rather than the upstream power supply, for better or for worse.
More of a concern is the shunt regulator only works up to a hard limit. If the instantaneous device current is greater than the sum of the static device current and the static shunt current, its game over. They are best used, therefore, in cases where the device current maximum can be accurately estimated so the shunt current can be matched to that number.
The basic circuit block can be upgraded with the inclusion of an error amplifier and feedback, just like a series regulator can. We'll try to have a look at that next time.
I've been meaning to take up shunt regulators for some time. I've never got around to building one myself to try, so I'll have to make do by playing in simulation.
Today's circuit is the shunt analog of the Z-reg series regulator: no feedback, Zener reference, single transistor regulation. The output impedance and ripple rejection-characteristics are similar too, with about 40 dB of RR and an output impedance of just a few ohms. It can be built equivalently from either an pnp or pnp transistor. (See attached LTSpice .asc files.)
The difference between shunt and series regulation can best be explained by considering the upstream power supply: In a series regulator an increase in current demand by the load causes the regulator to increase the current to compensate. In a shunt regulator an increase in current demand by the load causes the regulator to decrease the shunt current to balance, so there is no net change in current flowing out of the power supply.
If it all sounds too good to be true, it is! The main price is increased static current draw, though, which, for line level audio at least, isn't that much of an issue. The other point is the largish series resistor R1: its presence basically puts a brick wall between the circuit and the power supply, meaning the shunt regulator has to do all the heavy lifting in providing a low impedance power source to the amplifier circuit. In audio terms, you may hear the shunt regulator rather than the upstream power supply, for better or for worse.
More of a concern is the shunt regulator only works up to a hard limit. If the instantaneous device current is greater than the sum of the static device current and the static shunt current, its game over. They are best used, therefore, in cases where the device current maximum can be accurately estimated so the shunt current can be matched to that number.
The basic circuit block can be upgraded with the inclusion of an error amplifier and feedback, just like a series regulator can. We'll try to have a look at that next time.
Total Comments 1
Comments
-
is it the best result from all regulator on simulation ?Posted 14th January 2015 at 12:52 PM by oxoxbyx
